Data Wrangling with NRE Cleaned Cities

What does this notebook do?

It is the analisys of the cleaned NRW vergabe data.

Main conclusions:

Those tenders are the purchases made by public bodies that were not

0. Fixing the df

1.Distribution of tenders per month:

In the beggining of the system, we have a higher value os tenders added, but the amount of tenders go up and down during the period, beeing low at january 2022 partially due to lack of days (we got this data before 15.01.2022)

2. Proportion of types of tender:

According to the OCDE report, we should see at the beggining of the pandemic an increase of the non-competitive procurements. But as we see, both types behave the same way.

3. Information about sellers:

For all ex-post procedures, there is not a single record of result_value, result_seller_name, result_seller_town or result_seller_country

There is no way to know why these 4% of tenders have the data.

Without data of sellers, there is no way we can do monitoring to know, for example, if a company is not allowed to sell to the government

We have found only 92 inputs with the selected terms. Now we are going to look at them:

5. Who is buying?

6.What are they buying?

7.What are they buying?

8.When were they bought?

It doesn't say much.

9.What were the types of procedure used?

55% of the tenders were ex-post type.

Let's see if the ex-post ones were mostly by the beggining of the pandemic

We see that in May 2021 it was the highest rate of purchases made by ex-post procedures. The purchases made in this dataset are not for hospitals, but for government bodies. Particularly, I don't see why july 2021 would be the moment with the most ex-post procurements.